2 Comments

User's avatar
Daniel Evensen's avatar

The fact that the splits against RHP are basically a list of the players with the best OPS in the game is a sign of how overrated lefty/righty splits are in general, in my opinion. Speaking from the point of view of a baseball simulator (and check out my blog if you're interested), the more wild and interesting the lefty/righty split is for any given season, the higher the chances are that there are big sample size issues.

In "The Book," Tom Tango concluded that it took right handed hitters about 2,000 plate appearances against left handed pitchers before measured platoon splits could be considered reliable — which is, what, about 3 seasons worth? Similarly, Tango concluded that lefty hitters needed about 2,000 plate appearances against left handed pitchers before they could be reliable. In other words, it's really only the veteran hitters whose splits can be considered "reliable" or "meaningful" — and it just so happens that the splits over numerous years tend to revert to historical means.

Interestingly, Tango concluded that splits were more meaningful for pitchers at lower sample levels. My gut tells me that this was due to skewing in his data, as he was looking at a pool of pitchers dominated by LOOGY types who had extremely short careers. But I haven't really taken the time yet to dig into it.

Honestly, all the hoopla about lefty-righty splits being extremely important comes largely from Strat-O-Matic marketing, in my opinion. Yes, it does matter in real life — but it's also very hard to measure, especially since managers use players in a way that reflects perceived platoon advantages. It's really hard to measure how good a right handed batter is against left handed pitchers if he never starts against left handed pitchers.

Also — what I love about Baseball-Reference these days is how easy it is to get all of this info, even for older seasons. There's a lot of good research just waiting to be done.

Expand full comment
1 more comment...

No posts